Autonomous sense and guide machine learning system

Abstract

A system for generating a machine learning system to generate guidance information based on locations of objects is provided. The system accesses training data that includes training time-of-arrival (“TOA”) information of looks and guidance information for each look. The guidance information is based on a training collection of object locations. The TOA of a look represents, for each object location of a training collection of object locations, times between signals transmitted by transmitters and return signals received by receivers. The return signals represent signals reflected from an object at the object location. The system trains a machine learning system using the training data wherein the machine learning system inputs TOA information and outputs guidance information.

1 . A method performed by one or more computing systems to guide movement of a platform, the method comprising, for each of a plurality of intervals,
 receiving time-of-arrival (“TOA”) information derived from TOAs determined based on times between signals transmitted by transmitters and return signals received by receivers wherein a return signal is reflected from an observed object at an object location; and   determining guidance information by applying a machine learning system that inputs TOA information and outputs guidance information, the machine learning system being trained using training data that includes TOA information and guidance information.   
     
     
         2 . The method of  claim 1  wherein the TOA information is the TOAs of a look and the guidance information is a collection of object locations for each look. 
     
     
         3 . The method of  claim 1  wherein the TOA information is a collection of object locations corresponding to TOAs of a look and the guidance information is a guidance instruction. 
     
     
         4 . The method of  claim 1  wherein the TOA information is TOAs of a look and the guidance information is a guidance instruction. 
     
     
         5 . The method of  claim 1  wherein the machine learning system includes a first machine learning system that inputs TOAs of a look and outputs a collection of object locations and a second machine learning system that inputs a collection of object locations and output a guidance instruction. 
     
     
         6 . The method of  claim 1  wherein the platform is a component of a robot control system. 
     
     
         7 . The method of  claim 1  wherein the platform is a satellite and the object locations are locations of objects in space. 
     
     
         8 . The method of  claim 1  wherein the platform is an unmanned vehicle. 
     
     
         9 . The method of  claim 1  further comprising guiding the platform based on the guidance information. 
     
     
         10 . A method performed by one or more computing systems for evaluating an architecture of a sensor array for a platform, the method comprising:
 accessing a plurality of architectures of sensor arrays, an architecture specifying number and positions of transmitters and receivers of the sensor array; and   for each of plurality of architectures,
 for each of a plurality of time-of-arrivals (“TOAs”) of looks and an evaluation collection of object locations,
 applying a machine learning system that inputs the TOAs of the look and generates an estimated collection of object locations, the machine learning system trained using training data that includes TOAs of looks and for each look an evaluation collection of object locations; and 
 generating a metric based on similarity of object locations of the evaluation collection and the estimated collection; and 
 
 generating an architecture metric based on the metrics generated for the architecture. 
   
     
     
         11 . The method of  claim 10  wherein the architecture further specifies curvature of a sensor array. 
     
     
         12 . The method of  claim 10  wherein the architecture metric is further based on size or weight of a sensor array. 
     
     
         13 . The method of  claim 10  wherein the sensor array is a phased sensor array. 
     
     
         14 . The method of  claim 10  wherein an architecture further specifies a beam pattern of a transmitter. 
     
     
         15 . One or more computing systems to guide movement of a platform, the one or more computing systems comprising:
 one or more computer-readable storage mediums for storing computer-executable instructions for controlling the one or more computing systems to, for each of a plurality of intervals:
 receive time-of-arrival (“TOA”) information derived from TOAs determined based on times between signals transmitted by transmitters and return signals received by receivers; and 
 generate guidance information by applying a machine learning system that inputs TOA information and outputs guidance information, the machine learning system being trained using training data that includes TOA information and guidance information; and 
   one or more processors for executing the computer-executable instructions stored in the one or more computer-readable storage mediums.   
     
     
         16 . The one or more computing systems of  claim 15  wherein the TOA information is the TOAs of a look and the guidance information is a collection of object locations for each look. 
     
     
         17 . The one or more computing systems of  claim 15  wherein the TOA information is a collection of object locations corresponding to TOAs of a look and the guidance information is a guidance instruction. 
     
     
         18 . The one or more computing systems of  claim 15  wherein the TOA information is TOAs of a look and the guidance information is a guidance instruction. 
     
     
         19 . The one or more computing systems of  claim 15  wherein the machine learning system includes a first machine learning system that inputs TOAs of a look and outputs a collection of object locations and a second machine learning system that inputs a collection of object locations and output a guidance instruction. 
     
     
         20 . The one or more computing systems of  claim 15  wherein the platform is a component of a robot control system. 
     
     
         21 . The one or more computing systems of  claim 15  wherein the platform is a satellite and the object locations are locations of objects in space. 
     
     
         22 . The one or more computing systems of  claim 15  wherein the platform is an unmanned vehicle. 
     
     
         23 . The one or more computing systems of  claim 15  wherein the instructions further guide the platform based on the guidance information. 
     
     
         24 . One or more computing systems for evaluating an architecture of a sensor array for a platform, the one or more computing system comprising:
 one or more computer-readable storage mediums for storing computer-executable instructions for controlling the one or more computing systems to:
 access an architecture that specifies number and positions of transmitters and receivers of the sensor array; and 
 for each of a plurality of time-of-arrivals (“TOAs”) of looks and an evaluation collection of object locations,
 apply a machine learning system that inputs the TOAs of the look and generates an estimated collection of object locations; and 
 generate a metric based on similarity of object locations of the evaluation collection and the estimated collection; and 
 
 generate an architecture metric based on the metrics generated for the architecture; and 
   one or more processors for executing the computer-executable instructions stored in the one or more computer-readable storage mediums.   
     
     
         25 . The one or more computing systems of  claim 24  wherein the machine learning system is trained using training data that includes TOAs of looks and for each look an evaluation collection of object locations. 
     
     
         26 . The one or more computing systems of  claim 24  wherein the architecture further specifies curvature of a sensor array or a beam pattern of a transmitter. 
     
     
         27 . The one or more computing systems of  claim 24  wherein the architecture metric is further based on size or weight of a sensor array. 
     
     
         28 . The one or more computing systems of  claim 24  wherein the sensor array is a phased sensor array. 
     
     
         29 . One or more computing systems for generating a machine learning system to generate guidance information based on locations of objects, the one or more computing systems comprising:
 one or more computer-readable storage mediums for storing computer-executable instructions for controlling the one or more computing systems to:
 access training data that includes, for each of a plurality of looks, training time-of-arrival (“TOA”) information of the look and guidance information for the look; and 
 train a machine learning system using the training data wherein the machine learning system inputs TOA information and outputs guidance information; and 
   one or more processors for executing the computer-executable instructions stored in the one or more computer-readable storage mediums.   
     
     
         30 . The one or more computing systems of  claim 29  wherein the guidance information is based on a training collection of object locations and wherein the TOAs of a look represent, for each object location of a training collection of object locations, times between signals transmitted by transmitters and return signals received by receivers wherein the return signals represent signals reflected from an object at the object location. 
     
     
         31 . The one or more computing systems of  claim 30  wherein the training TOA information is generated during movement of a platform with transmitters and receivers through a volume of objects and object locations of the training collections are identified by an actual object location sensor. 
     
     
         32 . The one or more computing systems of  claim 29  wherein the TOA information represents the TOAs of a look and the guidance information is a collection of object locations for each look. 
     
     
         33 . The one or more computing systems of  claim 29  wherein the training TOA information represents a training collection of object locations corresponding to TOAs of a look and the guidance information is a guidance instruction. 
     
     
         34 . The one or more computing systems of  claim 29  wherein the training TOA information represents TOAs of a look and the guidance information is a guidance instruction.